INDEX

Explanations

words related to recognition or acknowledgement of achievements

oai_token-act-pair · gpt-3.5-turbo

terms related to recognition and achievement

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Flask

-0.73

CDC

-0.70

Cloud

-0.69

connection

-0.68

 Lazarus

-0.67

 Void

-0.66

adel

-0.65

 containment

-0.63

agic

-0.63

alg

-0.61

POSITIVE LOGITS

hon

4.47

Hon

1.77

hon

1.56

Hon

1.48

 hone

1.43

 honors

1.38

 honour

1.24

 Honour

1.19

 laure

1.02

 reve

1.00

Activations Density 0.015%

words related to recognition or acknowledgement of achievements

terms related to recognition and achievement

No Comments

No Known Activations